agentleFS
Jurel89Jurel89/copilot-omni/skills/configure-notifications/SKILL.md

configure-notifications

Skill0 starsChanged 7 months agoMITView on GitHub

Wire Telegram / Discord / Slack notifications for copilot-omni run events (start, progress, completion, error)

Files
SKILL.md
<Purpose> Let copilot-omni post notifications to Telegram, Discord, or Slack when run events fire. The skill is declarative: you give it a webhook (Discord, Slack) or a bot token + chat-id pair (Telegram), and it writes a config entry under .omni/config.json > notifications. The helper script scripts/notify.py reads that config and emits on stdin → HTTP POST. </Purpose>

<Use_When>

  • User asks for Slack / Discord / Telegram notifications.
  • User wants to know when a long autopilot run finishes.
  • User is setting up an ops on-call handoff for a repo-scoped automation (CI mirror, nightly scheduled skill). </Use_When>

<Do_Not_Use_When>

  • User wants email — not supported by this skill (v1.1 backlog).
  • User wants PagerDuty incidents — out of scope; integrate via Slack webhook + PagerDuty's Slack app.
  • User wants per-event templating beyond the four built-in event types (start|progress|done|error) — edit scripts/notify.py directly. </Do_Not_Use_When>

Configuration shape

Entries are persisted under .omni/config.json:

{
  "notifications": [
    { "target": "telegram",
      "bot_token": "123:ABC…",
      "chat_id": "-1001234567890",
      "events": ["done", "error"] },
    { "target": "slack",
      "webhook": "https://hooks.slack.com/services/…",
      "events": ["start", "progress", "done", "error"] },
    { "target": "discord",
      "webhook": "https://discord.com/api/webhooks/…",
      "events": ["error"] }
  ]
}
  • events defaults to ["done", "error"] when omitted.
  • Credentials never leave .omni/config.json; the hooks only read-through.

Usage

# Add a Telegram channel for done+error events
python3 scripts/notify.py configure telegram \
    --bot-token "$TELEGRAM_BOT_TOKEN" \
    --chat-id "$TELEGRAM_CHAT_ID" \
    --events done,error

# Emit a test event
python3 scripts/notify.py emit done "autopilot finished cleanly"

# List configured targets (credentials elided)
python3 scripts/notify.py list

Triggering from skills

Long-running skills (autopilot, ralph, ultrawork, team) SHOULD call scripts/notify.py emit <event> "<message>" at the matching lifecycle transitions. Failure to reach the remote service is non-fatal — notify.py logs the HTTP error to stderr and exits 0 so the main run is not blocked on a webhook outage.

See also

  • docs/STATE-MACHINE.md — the four-gate lifecycle that drives event emission points.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.